Unmanned Aerial Vehicle (UAV) Based Traffic Monitoring and Management
Total Page:16
File Type:pdf, Size:1020Kb
CAIT-UTC-NC8 Unmanned Aerial Vehicle (UAV) Based Traffic Monitoring and Management FINAL REPORT AUGUST 2016 Submitted by: Peter J. Jin Seyedamirali M. Ardestani Assistant Professor PhD Candidate Yizhou Wang Wangsu Hu PhD Candidate PhD Candidate Center for Advanced Infrastructure and Transportation (CAIT), Rutgers, The State University of New Jersey, Piscataway, NJ, 08854 External Project Manager Dhanesh Motiani Assistant Commiss io ner New Jersey Department of Transportation 1035 Parkway Ave Trenton, NJ 08625 In cooperation with Rutgers, The State University of New Jersey And New Jersey Department of Transportation Disclaimer Statement The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein. This document is disseminated under the sponsorship of the Department of Transportation, University Transportation Centers Program, in the interest of information exchange. The U.S. Government assumes no liability for the contents or use thereof. The Center for Advanced Infrastructure and Transportation (CAIT) is a National UTC Consortium led by Rutgers, The State University. Members of the consortium are the University of Delaware, Utah State University, Columbia University, New Jersey Institute of Technology, Princeton University, University of Texas at El Paso, Virginia Polytechnic Institute, and University of South Florida. The Center is funded by the U.S. Department of Transportation. 1. Report No. 2. Government Accession No. 3. Recipient’s Catalog No. CAIT-UTC-NC8 4. Title and Subtitle 5. Report Date Unmanned Aerial Vehicle (UAV) based Traffic August 2016 Monitoring and Management 6. Performing Organization Code CAIT/Rutgers University 8. Performing Organization Report No. TECHNICAL REPORT STANDARD TITLE PAGE 7. Author(s) CAIT-UTC-NC8 Peter J. Jin, Ph.D., Seyedamirali M. Ardestani, Yizhou Wang, Wangsu Hu 9. Performing Organization Name and Address 10. Work Unit No. Center for Advanced Infrastructure and Transportation (CAIT), Rutgers, The State University of New Jersey, 11. Contract or Grant No. Piscataway, NJ, 08854 437372 12. Sponsoring Agency Name and Address 13. Type of Report and Period Covered Center for Advanced Infrastructure and Transportation Final Report Rutgers, The State University of New Jersey July 2014-August 2016 100 Brett Road 14. Sponsoring Agency Code Piscataway, NJ 08854 15. Supplementary Notes U.S. Department of Transportation/OST-R 1200 New Jersey Avenue, SE Washington, DC 20590-0001 16. Abstract With recent Federal Aviation Administration (FAA) policy changes and test site designations, Unmanned Aerial Vehicles (UAVs) or “Drones” have gained attention among the public and private sectors. Practical applications such as facility inspection, mapping, surveillance, delivery, etc. have been intensively tested. Overt the last decade, UAV applications in transportation engineering have included experiments with traffic surveillance, infrastructure monitoring, and roadway incident management. Most have focused on transmitting on-site video footage to traffic management centers so traffic operators can monitor congestion, coordinate incident response crews, or collect traffic data in areas without CCTV surveillance systems. In this study, the main focus is to explore the feasibility of using UAV to accelerate the site surveying at major traffic accidents, a major delaying event in incident management. We developed a prototype UAV system that can be rapidly deployed in the field for video- based site surveying and 3D reconstruction of accident sites. The hardware system includes UAVs equipped with high-frequency GPS sensor, high-resolution camera, HD transmitter and ground station for communication and data collection. The software components include the mission control, planning software, and photogrammetry 3D reconstruction tools. The prototype system was tested at an orchestrated accident site and the developed 3D models that yield promising potentials for further development. The test results reveal factors such as shooting altitude and angle, glares on smooth surfaces, Geo-tagging of photo snapshots, and GPS signals have major impacts on the scanning results. These impacts need to be addressed efficiently to ensure the quality of the constructed model. 17. Key Words 18. Distribution Statement Unmanned Aerial Vehicle, Traffic Monitoring, Traffic Management 19. Security Classification (of this report) 20. Security Classification (of this page) 21. No. of Pages 22. Price Unclassified Unclassified Total #20 Form DOT F 1700.7 (8-69) Table of Contents 1. INTRODUCTION AND PROBLEM DEFENITION 2. LITERATURE REVIEW 3. CONCEPT OF OPERATION 4. EXPERIMENTAL DESIGN AND FINDIFDNGS 5. CONCLUSION AND RECOMMENDATION List of Figures 1. Locations of existing UAV studies on transportation infrastructure 2. Aircraft unit: (a) UAV (b) built-in GPS and transceiver (c) UAV during flight (d) batteries and camera gimbal 3. Mobile LiDAR Scanner and 3D points clouds infrastructure 4. Components: (a) camera gimbal components (b) Ground station components 5. Flight mission: (a) flight mission in software (b) flight mission top view in Google Earth (c) flight mission side view Google Earth 6. Images taken by UAV: (a) image location on Google Earth (b) sample accident site images from different angles 7. ST map generation 8. Schematics of LiDAR model Integration for Coordinate Transformation 9. Rutgers Mobile LiDAR System 10. Pixel Trajectory Generation (Dotted Lines are generated trajectories 11. Orchestrated accident site 12. 3D reconstruction results from Agisoft: (a) camera locations (b) close view 3D dense point clouds List of Tables 1. Image Coordinates 2. Measurement result 1. INTRODUCTION AND PROBLEM DEFINITION Traffic incident management (TIM) is the core operational mission in most traffic management centers (TMCs). Reducing incident response delay, ensuring the safety of response crews, completing thorough site investigation, and accelerating the incident recovery have been the main objectives in TIM operational improvement. Thorough site surveys and medical examinations at major accidents are the most critical but also the most time-consuming steps in the TIM process, but are necessary to facilitate subsequent safety analysis, medical treatment, insurance requirements, and legal proceedings. Meanwhile, capacity- restricting traffic accidents can cause large-scale traffic jams even in non-peak hours. Conducting comprehensive and expeditious accident site investigations are crucial to accelerate the process and to reduce the overall impact on traffic flow. In the state of New Jersey, major improvements have reduced the incident management duration for major accidents from 2.5 hours in 1995 to 40 minutes in 2014(1). However, the remaining road blocks to further reduce the duration is bounded by the time consumed on site through surveying by both law enforcement and medical crews as well as the clearance process. In this paper, the focus is to use Unmanned Aerial Vehicles (UAVs) to help reduce the duration of site surveying. In recent years, UAVs have shown high potential in remote sensing in a wide variety of areas. Embedded new technologies in UAVs enable them to fly convenient, fast, precise, safe and economical compared to other modes of remote sensing. The latest technologies are 1) a fully automatic flight mission including the waypoint execution and landing, 2) GPS-based position hold, 3) long-range wireless communication, and 4) long range data transmission. There are two types of UAVs, fixed-wing and rotary- wing. The former type is more suitable for large-scale sensing and surveying mission; while the latter one is more efficient for monitoring, surveillance, and surveying work that requires waypoint holdings and camera repositioning. The rotary-wing UAVs such as quadcopters can perform vertical take-offs and landings and conduct GPS or altitude holding in the air. These type of UAVs are more suitable for accident site investigation due to the limited geographical area of an accident site and the requirement on positioned photo or video shooting. Quadcopters are a type of UAVs with four rotary wings. Quadcopters are capable of Vertical Take-Off and Landing (VTOL), which gives more freedom for the pilot due to the fact that it requires much less space and time for landing and launching. Also, hovering capability of the Quadcopters provides an excellent condition for static remote sensing during flight. All these features could help the responders to 1) faster access to the accident site 2) prioritize the incident treatment 3) allocate fewer emergency resources, and 4) collect data faster which leads to faster site clearance. In this study, we propose the development of a prototype multi-functional airborne traffic management system (Air-TMS) for non-recurrent traffic congestion. The system includes Quadcopters Unmanned Aerial Vehicle (QUAV) equipped with a high-resolution video camera and HD video transmission units and the software to conduct waypoint planning and photogrammetry 3D reconstruction. A completely scaled 3D model of an orchestrated accident site is created and evaluated. An algorithm has been developed to detect and track congestion using LiDAR Model and traffic video. The prototype system has the potential of significantly accelerating the accident site investigation, with the entire UAV surveying process takes less than 10 minutes. 1 2. LITERATURE REVIEW Traffic Operation Studies Using UAV Studies have been done to